agentsclimarketplace

Planning

Skill cosmoscalibur/agent-harness/skills/planning

Produce an implementation plan before non-trivial code changes — scope, key design decisions, open ambiguities, and which decisions need developer approval vs. which are autonomous. Use before starting any non-trivial implementation, when a request could alter a public contract, persisted schema, or externally observable behavior, or when a task needs to be framed as a verifiable, test-first goal.From its SKILL.md

Install
npx -y skills add cosmoscalibur/agent-harness --skill planning

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

2 things to look at

  • no licenseNo license file was found in the repository. Code published without one is not open source by default, so using it at work is a question for whoever answers licensing questions where you are.
  • 1 stars1 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

SKILL.md

2.0 KB, as published. Nobody here has run it

Planning

Produce a plan before non-trivial implementation begins. Do not execute without explicit developer approval on the plan.

Rules

  • Before starting analysis, ensure the current branch is up to date with the remote default branch (fetch and compare, e.g. git fetch + git status -sb or git log HEAD..origin/<default>). A plan built on stale local state can target code that's already changed upstream.
  • The plan states: scope, key design decisions, open ambiguities.
  • Evaluate each design decision on the problem's actual nature — never default to a paradigm, pattern, or defense level without justifying fit; state that justification as one of the plan's key design decisions.
  • Flag which decisions need developer consultation (architecture trade-offs, contract changes, business-rule ambiguity) vs. which are autonomous (no impact on contract or observable behavior). Don't mislabel either direction.
  • Don't anticipate every internal function in the plan. Resolve unforeseen details autonomously per the implementation skill's resolution ladder; state the assumption inline.
  • Frame the task as a verifiable, test-first goal: "add validation" → write failing tests for invalid input, then pass them. "fix bug" → reproduce with a test, then fix. "refactor X" → tests green before and after.

Handoff

Once the plan is approved, proceed to the implementation skill. If execution has to diverge from the approved plan, that gets flagged explicitly during implementation — never applied silently.

Keep looking

Skills are one crate of 608,475. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.